import numpy as np
from sklearn.datasets import load_sample_images
from skimage import io
import cv2

# Visualisation libraries

## Text
from colorama import Fore, Back, Style
from IPython.display import Image, display, Markdown, Latex

## matplotlib
import matplotlib.pyplot as plt
from matplotlib.font_manager import FontProperties
from mpl_toolkits.axes_grid1.inset_locator import inset_axes
plt.style.use('seaborn-whitegrid')
import matplotlib as mpl
mpl.rcParams['figure.figsize'] = (17, 6)

%matplotlib inline

## seaborn
import seaborn as sns

import warnings
warnings.filterwarnings("ignore")

Getting Started with Open CV

We use the sklearn image dataset to demonstrate basic operations using Open CV.

Images = load_sample_images()
Names = [x.split("\\")[-1].replace('.jpg','').title() for x in Images['filenames']]
Images = Images['images']

def ImShow(Images, Names, Title = 'Images', axis_tight = False):
    fig, ax = plt.subplots(1, 2 , figsize = (17, 6))
    ax = ax.ravel()
    font = FontProperties()
    font.set_weight('bold')
    for i in range(2):
        _ = ax[i].imshow(Images[i])
        if axis_tight:
            _ = ax[i].axis('tight')
        _ = ax[i].axis('off')
        _ = ax[i].set_title(Names[i], fontproperties=font, fontsize = 16)
    _ = fig.subplots_adjust(wspace= 0.01)
    if Title:
        _ = fig.suptitle(Title, fontproperties=font, fontsize = 18)
    return fig, ax
# Image Show
_,_ = ImShow(Images, Names, Title = 'Original Images')

Accessing and Modifying pixel values

We can use split funtion to access various channels of an image.

B, G, R = cv2.split(Images[0])

Alternatively, we could use, for example, use Img[:,:,0], Img[:,:,1] and Img[:,:,2] to access the red, green and blue chanells of Img.

fig, ax = plt.subplots(1, 3 , figsize = (17, 3.5))
ax = ax.ravel()
font = FontProperties()
font.set_weight('bold')
_ = fig.suptitle(Names[0], fontproperties=font, fontsize = 18)

Channels = ['Red', 'Green', 'Blue']

for i in range(3):
    img = 0*Images[0]
    img[:,:,i] = Images[0][:,:,i]
    _ = ax[i].imshow(img)
    _ = ax[i].set_aspect('auto')
    _ = ax[i].axis('off')
    _ = ax[i].text(x = 0, y = 0, s = Channels[i], size=14,
                   color = 'Navy', bbox=dict(boxstyle="square", ec='Navy', fc='LightSkyBlue'))

Adding Pading

• cv2.BORDER_CONSTANT: Adds a solid colored border.
• cv2.BORDER_REFLECT: Border will be mirror reflection of the border elements, like this : fedcba|abcdefgh|hgfedcb
• cv2.BORDER_REFLECT_101 or cv2.BORDER_DEFAULT: Border will be mirror reflection of the border elements, like this : like this : * gfedcb|abcdefgh|gfedcba
• cv2.BORDER_REPLICATE - Last element is replicated throughout, like this: aaaaaa|abcdefgh|hhhhhhh
• cv2.BORDER_WRAP - It will look like this : cdefgh|abcdefgh|abcdefg

For more information, please see the offical documentation here.

Dict = {cv2.BORDER_CONSTANT:'BORDER_CONSTANT',
        cv2.BORDER_REPLICATE:'BORDER_REPLICATE',
        cv2.BORDER_REFLECT:'BORDER_REFLECT',
        cv2.BORDER_WRAP:'BORDER_WRAP',
        cv2.BORDER_REFLECT_101:'BORDER_REFLECT_101'}

# RGB Colors
RGB = [[0,0,0] for i in range(3)]
for i in range(3):
    RGB[i][i] += 255

fig, ax = plt.subplots(2, 3 , figsize = (17, 7))
ax = ax.ravel()
fig.delaxes(ax[-1]) 

font = FontProperties()
font.set_weight('bold')
_ = fig.suptitle(Names[1], fontproperties=font, fontsize = 18)

for i in range(len(Dict)):
    img = cv2.copyMakeBorder(Images[1], 15, 15, 15, 15, i, value= RGB[0])
    _ = ax[i].imshow(img)
    _ = ax[i].set_aspect('auto')
    _ = ax[i].axis('off')
    _ = ax[i].set_title( Dict[i], fontproperties=font, fontsize = 12)

Image Blending

Image blending can be done using cv2.addWeighted. This function blends two images as follows

$$\text{Output Image} = \alpha\,\text{Image1} + (1-\alpha)\, \text{Image2} + \gamma$$

where $0\leq \alpha\leq 1$ and $\gamma$ is a an integer.

Augmented_Images = Images.copy()
# First Method
Augmented_Images[0] = cv2.addWeighted(Images[0],0.7, Images[1],0.3, 0)
# Second Method
Augmented_Images[1] = np.round(((0.7)*Images[0] + (0.3)*Images[1])).astype('uint8') 

fig, ax = plt.subplots(1, 2 , figsize = (17, 6))
ax = ax.ravel()

font = FontProperties()
font.set_weight('bold')
_ = fig.suptitle('Image Blending', fontproperties=font, fontsize = 18)

Title = ['cv2.addWeighted(Images[0],0.7, Images[1],0.3, 0)', 'Our Function']
for i in range(2):
    _ = ax[i].imshow(Augmented_Images[i])
    _ = ax[i].set_aspect('auto')
    _ = ax[i].axis('off')
    _ = ax[i].set_title(Title[i], fontproperties=font, fontsize = 12)

Bitwise Operations

This includes bitwise AND, OR, NOT and XOR operations.

• bitwise_and: Calculates the per-element bit-wise conjunction of two arrays or an array and a scalar.
• bitwise_not: Calculates the per-element Inverts every bit of an array.
• bitwise_or: Calculates the per-element bit-wise disjunction of two arrays or an array and a scalar.
• bitwise_xor: Calculates the per-element bit-wise “exclusive or” operation on two arrays or an array and a scalar.

Git = io.imread('https://github.githubassets.com/images/modules/open_graph/github-mark.png')
scale = 0.05 # of original size
Git_resized = cv2.resize(Git, tuple(np.dot(scale,Git.shape[:2][::-1]).astype(int)), interpolation = cv2.INTER_AREA)

## Invert
# Git_resized =  255 - Git_resized
Git_resized = cv2.bitwise_not(Git_resized)

Img = Images[0]
Mask = 0*Img
Mask[0: Git_resized.shape[0], 0:Git_resized.shape[1],:] = Git_resized
Img = cv2.add(Img,Mask)
fig, ax = plt.subplots(1, 2 , figsize = (17, 6))
_ = ax[0].imshow(Mask)
_ = ax[1].imshow(Img)

for i in range(len(ax)):
    _ = ax[i].set_aspect('auto')
    _ = ax[i].axis('off')

_ = ax[0].set_title('Logo', fontproperties=font, fontsize = 12)
_ = ax[1].set_title('Image + Logo', fontproperties=font, fontsize = 12)

Changing Colorspaces

This is done via cvtColor which converts an image from one color space to another.

RGB to Grayscale

Augmented_Images = Images.copy()
for i in range(len(Augmented_Images)):
    Augmented_Images[i] = cv2.cvtColor(Augmented_Images[i],cv2.COLOR_BGR2GRAY)

fig, ax = plt.subplots(1, 2 , figsize = (17, 6))
ax = ax.ravel()
font = FontProperties()
font.set_weight('bold')
for i in range(2):
    _ = ax[i].imshow(Augmented_Images[i],'gray')
    _ = ax[i].axis('tight')
    _ = ax[i].axis('off')
    _ = ax[i].set_title(Names[i], fontproperties=font, fontsize = 16)
_ = fig.subplots_adjust(wspace= 0.01)
_ = fig.suptitle('RGB to Grayscale', fontproperties=font, fontsize = 18)
        

Augmented_Images = Images.copy()
for i in range(len(Augmented_Images)):
    Augmented_Images[i] = cv2.bitwise_not(cv2.cvtColor(Augmented_Images[i],cv2.COLOR_BGR2GRAY))

fig, ax = plt.subplots(1, 2 , figsize = (17, 6))
ax = ax.ravel()
font = FontProperties()
font.set_weight('bold')
for i in range(2):
    _ = ax[i].imshow(Augmented_Images[i],'gray')
    _ = ax[i].axis('tight')
    _ = ax[i].axis('off')
    _ = ax[i].set_title(Names[i], fontproperties=font, fontsize = 16)
_ = fig.subplots_adjust(wspace= 0.01)
_ = fig.suptitle('RGB to Grayscale (Inverted)', fontproperties=font, fontsize = 18)

RGB to HSV

Augmented_Images = Images.copy()
for i in range(len(Augmented_Images)):
    Augmented_Images[i] = cv2.cvtColor(Augmented_Images[i],cv2.COLOR_BGR2HSV)
_,_ = ImShow(Augmented_Images, Names, Title = 'RGB to HSV')

---

Refrences

1. OpenCV documentation

---